Longitudinal patterning of depression repeatedly assessed across time among youth: Different trajectories in self-report questionnaires and diagnostic interviews.

Developmental epidemiological work shows that rates of depression as assessed by diagnostic interviews increase from childhood through early adulthood. It could be assumed that the trajectory of depression as assessed by self-report questionnaire measures would be characterized by a similar pattern. We aimed to evaluate this assumption and more clearly establish the longitudinal trajectory of depression in youth, when repeatedly assessed over time with a self-report questionnaire and with a diagnostic interview. Participants were 679 youth ages 7-16 years at baseline (M-age = 11.8, SD = 2.4, 56% girls). They completed the Children's Depression Inventory (CDI) every 3 months for 3 years (13 time points) and were interviewed every 6 months using the Kiddie Schedule for Affective Disorders and Schizophrenia (K-SADS) to ascertain onset of depression diagnosis. A series of growth curve models was fit to the CDI and K-SADS data. A piecewise model characterized growth in depression as assessed by the CDI, with an initial negative linear slope (b = -0.64) spanning the first 3 assessments, and a positive quadratic second slope (b = 0.015; linear component: b = -0.22) spanning the remaining 10 assessments. Depression, as assessed by the K-SADS, grew continuously over time (a positive linear slope, h = 0.23). Findings illustrate differences between longitudinal trajectories of depression when assessed repeatedly by self-report questionnaire and diagnostic interview. Implications for research designed to study longitudinal depression trajectories are discussed.